Cowardlyducks - E-BikeE builds

Ok, so I looked into this Galvanic corrosion thing a bit more, and I 'think' I should be ok.

According to the below link, because I've used Zinc fasteners, corrosion is not increased on the aluminium surface of the frame.
http://www.preservationscience.com/materials/metals/PGC.html
galvanicchart.gif

The corrosion of the zinc faster against the steel is increased, but because I plan to replace these fasteners within the next 6 months, I don't see that being a huge problem. That surface is on the outside anyway, so I should be able to see anything obviously wrong.

For the Stainless Steel bolts I've ordered corrosion is marginally increased. Given it's only marginal, I'm ok with that as I almost never ride in the rain.
When I swap over to the stainless bolts, I might throw some grease or petroleum jelly on there to also help prevent water ingress.

Cheers
 
EDIT: Ignore the below, I totally missed the picture showing separate bolts per side. :oops:

[strike]Question, as I am not sure from the pics/info above:

Did the original rivets go all the way thru both sides of the bracket and frame, or were there separate sets of rivets for left and right side?

If the latter, then on the bolts you're using, do you have a tube over each bolt, inside the frame, to prevent potential distortion/crushing of the bracket and frame over time? Or is this not a possible issue due to the way loads occur on that interface?[/strike]
 
Thanks for the interest guy's.
I finally got everything wired back up and put back together tonight. It takes most of my time working on this bike just disassembling and reassembling everything as there are quite a lot of little fiddly bits all over the place.

Thinking ahead to my new battery, I would like to reduce the number of cables running from my handle bars to the rear of the bike.
I'm currently using a Cat5 network cable for my CA, and an old DVI cable for my throttle, 3 speed switch, cruise, on/off, and regen wires.

I would like to be able to combine these 2 wires into one if possible, but I'm having a hard time finding anything suitable.

Does anyone know of any 18+ core wires no larger than 3.5mm diameter?

Cheers
 
I used an old thin "low quality" parallel printer cable to get a bunch of tiny wires in one cable--I didnt' need connectors so I just cut off both ends and used the cable itself, soldering to everything at each end. It was thicker than 3.5mm, though, probably closer to 6, including the insulation of the outer jacket. Maybe without the jacket it would've been thin enough for you. Or you could slit it and pull out some of the wires, as there's up to 26 of them (if it has a shield ground), and cut away just part of the jacket lengthwise along the slit to reduce the diameter.

VGA cables (especially cheap ones) can often be down at or close to that size, though they'll only have at most 16 wires (if there is a shield ground), and the cheap ones may not even have all the wires.
 
So I just bought 6 of these:
56845.jpg

I thought I would never buy more LiPo, but these were too cheap to pass up. They were nearly half price at $65 Aud each.

I still plan to do a frame pack eventually, but for now, 4 of these in a 2S2P setup will go on the recumbent to get me through the winter. It's not even that cold yet and I'm hitting 3.5V per cell as I get home most day's. :|

The weight will be a little less than the current setup at around 4.75Kg vs the current 5.03Kg.

I bought 6 so I can use them on my Fighter also. With 6 in the Fighter I will have 30Ah, so I might not get/build both packs at the same time any more and just start with the recumbent pack so I can move these multistars over to the Fighter and stop using that piece of crap LiFeP04 pack that's sagging so bad. :roll:

Sorry to disappoint those who were looking forward to my frame pack soon.

Cheers
 
about aluminum vs. stainless hardware:
I used 18-8 SS hardware and aluminum extrusions to construct these solar panel frames 10 years ago. They still show very little corrosion and I live next to an ocean in a rain forest i.e they get really wet really often.
The zinc and steel hardware I initially used for hinge pins didn't last 2 years before completely failing/rusting to dust.
Did I mention that SS pins are absurdly expensive? They are.
steel pins cost $5; stainless steel pins cost $60side show.jpgstainless type.jpg
 
Well, the MultiStars are on the bike. Weight is a little lower, and range should be longer while I have them on, but the assembled padded bag they are in is basically the same size as the previous setup, so no visual change to report in that regard really.
I put through an order with Allex for 12 x spot-welded 5P 18650 'bricks' with tabs using Sanyo GA 3500mah cells. Not sure when I'll get them as he's quite busy, but once received I'll still have to solder up the series links, assemble and pad it all out, and sort out all the balance wires, so it's still a little while off for this bike.

Apart from that I've been working on permanently (hopefully) repairing the corners of the seat-back fabric which keeps tearing. I unpicked the half dozen or so 'quick' repair jobs I've done over the years and this was the end result:
DSC_3078.jpg

DSC_3080.jpg


So my wife kindly donated some old jeans for the cause and I will be sewing some patches from them into the inside of the corners:
DSC_3076.jpg

DSC_3082.jpg


Hopefully this sorts it out once for all. :roll:

Cheers
 
I finished off the work on the seat-back last night and took it for a spin today to see a local car show, and had a blast in Kiwiev's electric Sonic 7. :mrgreen:

I'm not sure exactly what changed, but it seems a lot more comfortable now. It's nice also not having to worry when I lean back about it tearing through the tops of the posts.
Anyway, i'll let the pics tell the rest of the story:
DSC_3129.jpg
DSC_3130.jpg

DSC_3128.jpg
DSC_3131.jpg

I also added this compression band normally used for back pain or pregnancy to better hold the foam padding in place inside the seat-back.
DSC_3132.jpg
DSC_3134.jpg

The black bands top and bottom are actually elastic cotton head sweat bands. They aide in holding everything in place and because they're elastic also help support my back.
DSC_3136.jpg
DSC_3139.jpg


Also, Allex has finished building the parallel bricks for my frame pack. Once I receive them, I can start assembling them into a long pack and adding series links. :)
cowardlyduck.jpg

I plan to add 2X series wires soldered across the top and bottom of the length ways nickel strips of each of the parallel bricks. Not sure if I'll use 14awg or 12awg yet...depends on how much spare space I have top/bottom to some degree.

Cheers
 
So DAM!
Looks like I won't be getting that battery Allex made for me after all. :(
After a few attempts at posting it by different means it kept getting returned to him, the last time by air with a note saying not to ship batteries by air. :roll:
https://endless-sphere.com/forums/viewtopic.php?f=6&t=23996&p=1195567#p1195567

Allex has been great through the whole ordeal, keeping me updated and doing his best to make it right, but in the end it just wasn't happening so my money was returned.

Anyway, I think I will now just buy the cells and have to make the pack myself. So I'll see if Tumich is able to ship to me, or if that doesn't work, will try some local sellers.

Cheers
 
Been commuting with the recumbent most day's lately after a few weeks of barely riding it due to rain.

I think I may have mentioned it before, but the Infineon controllers seem to have a bug.

Well today I was playing around a bit with this bug as I was riding home, and decided to flick it right as I turned up a steep hill and the bike had slowed down to less than 25kph. Well this was the result:
DSC_3270.jpg

I could barely believe it when I got home and saw the stats. Remember, I'm running this bike on a 6FET 4110 Infineon.
The small 20" wheel certainly helps, and would only have been very momentary.
What's most amazing to me is that it was just 2S2P (12S 20AH) of these that did it:
56845.jpg

They barely broke a sweat, and the voltage sag was only about 4V. That gives me confidence in my planned 18S setup I'm going to use these for on my Fighter soon.

Cheers
 
So I bought an MPPT solar charger.
http://www.ebay.com.au/itm/141798591488
$_12.JPG

Given it's price, I'm highly sceptical and think it's probably just a DC-DC, but I thought I would give it a try anyway.

While I was modifying things, I went ahead and replaced all my 2.5mm 2pin JST connectors with 2mm HXT bullets and 18AWG wire:
HXT2MM.gif

These provide a much better connection and the step up from 20awg to 18awg should ensure minimal wiring losses. I've still got a long run of 20awg down the length of the bike I need to replace, but it was a huge PITA to get in there so I won't bother doing that until I move ahead with my frame battery and have to redo all the wiring anyway.
I also removed one of the diodes I had installed. I was using one before the DC-DC and one after, but this just seemed to rob me of the minimal power the setup had available, so I've just left the diode on the battery side for now.

While I was at it, I trimmed off the extra white space from the sides of the panel...not exactly sure why, but I had my scissors in my hand with it right in front of me and I just started cutting. :lol: I'd been thinking about doing it for a while, and am kind of glad I did as it makes the bike a bit less cumbersome. As you can see from the photo's below, I park it in my laundry next to a door, so not having the extra inch of solar panel each side actually makes it a good deal easier to park the bike and squeeze past it. :)
DSC_3272.jpg

DSC_3273.jpg

DSC_3276.jpg

DSC_3278.jpg

The MPPT has the added advantage of looking like it really could be my GPS tracking system. :lol:

I haven't got any stat's on whether or not it's improved the solar efficiency yet as the only day I've ridden it since making the changes was overcast and foggy all day. Now they're forecasting rain the rest of the week so I probably won't see if it's any better till next week. :x

It shouldn't be hard to beat the old setup though. In winter I was only seeing 0.4-0.9W charge typically while riding into the sun. If anyone has any other suggestions for improving the solar efficiency I'm all ears. :D

Cheers
 
eTrike said:
Test the panel with and without shading from handlebar area?
Can't really change that unfortunately. I originally did mount the panel lower, however the tops of my feet were hitting the bottom of the panel. It's current position is about as low as I can go without it becoming problematic unfortunately.
My only other option is to reposition the top part of the panel out from the handlebar area....something I might have to think about. I can see it might make the angle steeper which could reduce the overall efficiency.

Cheers
 
Huge thanks to ES member 9lives who is letting me borrow his JP spot welder over the coming months to put together my 18650 battery pack for this bike.

I'm slowly getting everything ready. Last night I put together this little 6 cell charger/discharger from 2 x 3cell holders. This should let me quickly and easily check through each cell before making the pack.
DSC_3558.jpg

DSC_3556.jpg


And this is my work area for spot welding the pack together. The corner of the dining table ain't great, but you gotta work with what you've got sometimes. Many a good creation has come out of this corner.
DSC_3559.jpg


Now to start cell sorting. :)

Cheers
 
Made my first 5P block of cells tonight.
This was a trial run using the spot welder to get a feel for doing it before I go ahead with the good quality Sanyo's.
DSC_3590.jpg

DSC_3589.jpg

DSC_3588.jpg


I'm not sure on the best approach for linking the two rows of cells. Am I better with 4 zig-zag links, linking every cell with every other cell like on the negative above. 3 links, bridging to the middle of the shorter row as I did on the positive side above. Or do I really not need that many links since it's just parallel balancing and 2 links bridging the centre of the row of 3 to the centre of the row of 2 would be enough?
I would be inclined to go with the latter, but I'm a bit uneasy about spot welding above the cell casing even though there's another layer of nickle beneath. Do I really not need to be concerned?
I might ask these same questions over on the 18650 spot welding thread.

Next up is a holiday, but I'll be getting back into this build in a couple of weeks. :)

Cheers
 
You may find wider or that train track pattern nickel strip is easier and faster and neater to weld with.

Also if be careful with the sharp edges hanging over the + button. A little knock and they can tip down inside and short. Being neat is the way to go.

Rectangle shaped bricks makes this easier than the trapezoid shape you have there. Either way good luck with it.

What pack size are you doing?

Sent from my UMI_SUPER using Tapatalk
 
Lash said:
You may find wider or that train track pattern nickel strip is easier and faster and neater to weld with.

Also if be careful with the sharp edges hanging over the + button. A little knock and they can tip down inside and short. Being neat is the way to go.

Rectangle shaped bricks makes this easier than the trapezoid shape you have there. Either way good luck with it.

What pack size are you doing?

Sent from my UMI_SUPER using Tapatalk
Thanks Lash.
Using any different nickel is out of the question now I just ordered 1KG of 7mm wide 0.15. Pretty sure I will never run out now. :roll:
Thanks for the tips on the sharp edges. It can't be seen, but I did actually round of the corners for the strips laid down on the bottom layer of the + side. I can see why it would also make sense to do that for any top layers also so will do that from now on.

The plan is to make one long pack approx 580 x 72 x 35mm (L x H x W) which is the largest it can really be made to still fit comfortably inside the frame with padding. I would love to be able to fit one extra parallel cell to make 6P, but it would mean splitting the pack into awkward segments in from of and behind the rear fork which I'm not keen on.
It will be 12S 5P, good for 16.5AH/ 730Wh real world usable which is just enough for my needs where I occasionally do 50-60km round trips for work, but normally do 20-30km trips. The lower weight compared to my current battery will also help somewhat with range and I plan on removing some extra unnecessary things from the bike to further reduce weight soon.

I also plan to eventually change controllers over to a Phase runner, and was thinking about changing motors to one of the newer 90% efficient motors. Those 2 combined with the lower weight battery should net me an extra 5-8% total system efficiency meaning I don't really miss the extra battery capacity.

And then there's the extra solar panels I've been thinking about. I learned recently that amorphous solar cells work well in low light conditions where mono crystalline can't produce much power. I've been thinking about lining the sides/top of the main frame beam with small amorphous solar cells to help boost solar charging performance. I need to do more research on if this is in fact true as there seems to be a lot of conflicting info out there.

Cheers
 
My pack build is coming along nicely.
DSC_0001.jpg


I had a bit of a close call when spot welding on the first block of 5 parallel cells, but the rest have been fine now after adjusting my approach. Details here:
https://endless-sphere.com/forums/viewtopic.php?f=14&t=68005&p=1253775#p1253775

Next is the painful job of soldering wire across the tops of these blocks. I fear it will be more difficult than the spot welding as getting the wire/solder hot enough to create good soldered joins is difficult to do with heat sucking cells.
I had planned on using 14awg fine stranded silicone wire, but after test soldering some across the top of my test 18650 5P cells, I didn't like how hot I had to get it. Testing with 16awg fine stranded worked much better and that's what I plan to use now. I still had to get it very hot, but I plan on heating it up with a butane torch before touching it to the pre-tinned nickel. I also tested this and it worked much better than holding the soldering iron to it and waiting for heat to build up...even with the soldering station set to 480C it wasn't hot enough to do the soldering fast enough for my liking.
I also wrap the cells in wet wipes to try and prevent heat getting to them, but really have no idea if this does anything.

Any tips people have on avoiding heat transfer to the cells would be appreciated. :)

16awg wire has a max recommended amps load of 20-25 amps, however I plan on using 2 of them in parallel across both nickel strip lines so in theory I should be good up to 50A, which I have peaked at before, but never maintain that kind of power continuously, so I think it should be fine.
I also plan on padding this pack with thermal padding as an added way of dissipating heat to the bike frame. Just not sure if I should do a layer of heat shrink over the cells first, then another layer including the thermal padding, or just throw the thermal padding directly onto the cells, then one layer of heat shrink over the whole lot. I've barely got any room for padding as it is.

Cheers
 
Why solder? Why dont you spot weld 4 x 0.15 nickel strips between the parallel groups? That will be OK for 50A peaks. You could spot weld them in an X pattern or just one on top of the other. Just be generous with the number of spots. I usually hit it 3 times so 6 spots. My latest Beta packs are doing 80A peaks with 5 X 0.15 nickel across the parallel groups. I only pull big amps for short periods of time. Nothing even gets warm.
 

Attachments

  • 20161202_162355.jpg
    20161202_162355.jpg
    106.7 KB · Views: 3,778
Kepler said:
Why solder? Why dont you spot weld 4 x 0.15 nickel strips between the parallel groups? That will be OK for 50A peaks. You could spot weld them in an X pattern or just one on top of the other. Just be generous with the number of spots. I usually hit it 3 times so 6 spots. My latest Beta packs are doing 80A peaks with 5 X 0.15 nickel across the parallel groups. I only pull big amps for short periods of time. Nothing even gets warm.
Thanks for the input John. That's a totally valid point I hadn't really considered that much.

I didn't think multiple layers of nickel was that good...even with multiple spots. At least Flippy doesn't seem to think so:
https://endless-sphere.com/forums/viewtopic.php?f=14&t=68005&p=1253775#p1250091

I've already been doing at least 6 spots on both the first welds and the second layer. So would you say just 2 more layers over the top of each row, each with 6 spots per cell would be ok?

Or what about a hybrid of both?

I could do 1 layer of joining nickel per row (so 2 total per parallel block), with 20awg wire soldered to the top of it? 20awg wire would be super easy to solder without over heating anything.

Cheers
 
I don't personally agree with spot welding a second layer is ineffective due to contact area. How is it different to the contact of the first layer on top of the cell?
You could hybrid if you like. Spot welding will give the pack rigidity also.
 
So in an effort to get the best of both worlds I did 2 layers of nickel + 18awg wire. Probably overkill but I can't easily remove this pack once it's installed so want to make it as solid as possible.
DSC_3642.jpg

DSC_3643.jpg

This will be a 2S add-on section to make my 12S into 14S and will sit separately behind the rear stay mounts.

I think I might just go with 20awg (or maybe even 24awg as that will still take a 10A load) for the rest as even the 18awg was pretty difficult to solder. The first one/two cells are pretty easy, but then they just suck the heat away making it increasingly difficult to solder the rest. I think it's also due to me using the high strand count silicone RC wire which absorbs heat far faster....but it also handles far higher currents. :)
I ended up super heating my soldering iron with my butane torch to get it hot enough to do the job quickly as I really hate holding the hot soldering iron to the wire waiting for it to melt into the pre-tinned nickel. I can feel the cells getting hot and I really don't like it, where as a super heated iron tip melts everything together quickly leaving the cells just a bit warm.

Cheers
 
The 2S addon pack is complete. :)

For the main discharge connectors I used 2 x 14awg for each pos/neg connection and 20awg for the balance wires.
I didn't take any photo's of the assembly but will try and get some when I do the main pack.

DSC_3646.jpg

DSC_3647.jpg


The 14awg wire was a massive PITA to solder to the cells/nickel. The cells got way hotter than I was comfortable with and at one point I was wondering if one of them was going into thermal runaway as it stayed hot for ages. I didn't measure temps, but a few times they were only just touchable without burning my fingers, so I'm guessing about 60C, but only for a few seconds.
Using 14awg was probably a bit overkill given I'm using 2 in parallel. I might go with 16awg for the main pack, but not sure as one of the connections will have to run the full length of the pack which is nearly 600mm.

As you can see above, I encased the pack into PVC heat shrink. Before I sealed it up, I cut thermal heat padding to size for the tops and sides and slid the heat shrink over it (with difficulty). If there's room still in the frame channel I will add additional padding, but I doubt I will fit much if any.

Something that I've realised I didn't consider with this extra 2S pack is how I will keep it in place. The main 12S pack will be able to sit against the frame bottom (with padding), however there will be nothing supporting the 2S pack from underneath. I could use zip ties like I do with my controller, but a more elegant solution that doesn't require screwing with the frame would be better. Any suggestions?

As a reminder, it looks like this when the bikes upside down:
DSC_2875.jpg'

This pack will sit just behind the rear most bolts, which you can see the hole for one on the right side of the frame.

Cheers
 
Back
Top